Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 20 Apr 1999 13:53:22 -0700 (PDT)
From:      Doug White <dwhite@resnet.uoregon.edu>
To:        Stanley.Hopcroft@ipaustralia.gov.au
Cc:        freebsd-questions@FreeBSD.ORG, Carl.Makin@ipaustralia.gov.au
Subject:   Re: 2.2.8-STABLE panic (page fault)
Message-ID:  <Pine.BSF.4.03.9904201349540.1514-100000@resnet.uoregon.edu>
In-Reply-To: <4A256758.001E911C.00@noteshub01.aipo.gov.au>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 19 Apr 1999 Stanley.Hopcroft@ipaustralia.gov.au wrote:

> supervisor write, page not prsent
> fault 12 page fault while in kernel
> 
> instruction pointer ?         0x08 0xf0 13 b2 4b
> stack pointer            0x10 0xef bf fe d8
> frame pointer            0x10 0xef bf ff e0
> code segment             base 0x0 limit 0xfffff
>                     type 0x1b DPL 0 pre 1 def 32
> 
> processor edlags         interuptable resume IOPL = 0
> current process               httpd
> page fault

You've attempted to recreate the message and have left out important info.
Please recopy the data and post it EXACTLY as printed.  This is _very_
important!

>  It has been very stable while running 2.2.7-RELEASE. This is now the
> second time I have had a 2.2.8-STABLE machine fail (can't say what
> happened in an other machine with identical processing load and very
> similar hardware) in this role.

> 
> The last message in the log before the reboot messages was "out of mbuf
> clusters. Increase maxusers!

You need to crank up the number of mbufs.  Try adding

options "NMBCLUSTERS=8192"

to your kernel config and rebuilding.  

The dead ringer for this sort of thing is netstat -m output:

sturbei,ttyp3,~,24>netstat -m

62/256 mbufs in use:
        33 mbufs allocated to data
        29 mbufs allocated to packet headers
32/82/1536 mbuf clusters in use (current/peak/max)
      ^^^^ you care about this number

196 Kbytes allocated to network (36% in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

If the mbuf clusters get to within 1/3 of the system max (4096?) then you
need the NMBCLUSTERS option.

See LINT and the mailing list archives for details.

Doug White                               
Internet:  dwhite@resnet.uoregon.edu    | FreeBSD: The Power to Serve
http://gladstone.uoregon.edu/~dwhite    | www.freebsd.org



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.03.9904201349540.1514-100000>